Dark Exchange: a 100% decentralized p2p exchange

[finway]

How do you decnetralize  US Dollars ?

[jtimon]

How do you decnetralize  US Dollars ?

The trades are not atomic nor automatic. Both parties are responsible for not being scammed in the trade.
But the trades could be atomic (when the bitcoin protocol allows replacements) for other currency pairs (say a btc/nmc trade) and could be automated within dark exchange.
To decentralize dollars, you could make a coin backed by USDs:

https://bitcointalk.org/index.php?topic=9493

But of course, then you would need to trust the usdCoin issuer instead of an exchange manager.

[morpheus]

Unfortunately, I haven't had time recently to put much work into Dark Exchange. I was in the middle of implementing a trust score when things got crazy at work.

Since there still seems to be interest in the concept, I'll have to put more time into it. Of course, I'm always open to help.

[Technomage]

There is definitely interest, keep up the good work! I made a donation to support this project.

This is actually quite significant. If we can use a dark exchange system that has good web of trust attributions, it will massively reduce the risk of centralized exchanges, such as Mt. Gox.

We wouldn't need to worry *that* much about the exchanges even though them going down would make it much more difficult for Bitcoin to go any more mainstream. But for power users a dark exchange is an excellent backup and for some of us it might even be preferable.

[cbeast]

This must be a very important idea, but I don't have a clue how it's supposed to work. :/

[Sultan]

This must be a very important idea, but I don't have a clue how it's supposed to work. :/

I'm a bit confused too :-(

I am guessing the decentralised nature of the exchange is with regards to the bids and asks alone, but as for the actual trade, that would probably be out of scope and dependant on the trust between the individuals on a trade.

Nevertheless, a decentralised database for, essentially storing the price of Bitcoin is crucial.

All what I said above is merely assumptions, I don't think I understand it.

[jtimon]

I am guessing the decentralised nature of the exchange is with regards to the bids and asks alone, but as for the actual trade, that would probably be out of scope and dependant on the trust between the individuals on a trade.

I haven't installed or read the software, but it's how it works as far as I can tell.

Nevertheless, a decentralised database for, essentially storing the price of Bitcoin is crucial.

I didn't thought about this in this way. Thank you.
This sentence has really helped me.

[cbeast]

I am guessing the decentralised nature of the exchange is with regards to the bids and asks alone, but as for the actual trade, that would probably be out of scope and dependant on the trust between the individuals on a trade.

I haven't installed or read the software, but it's how it works as far as I can tell.

Nevertheless, a decentralised database for, essentially storing the price of Bitcoin is crucial.

I didn't thought about this in this way. Thank you.
This sentence has really helped me.

I wrote this in another thread earlier today.

It's not like we need a full exchange that reports the volume of transactions. All we need is a general idea of the price bitcoin is being traded at.

• An Exchange Coin is created. The coins generated from it include pricing of all included crypto-currencies. More can be added if everyone agrees.
  
• Each coin client (including bitcoin) allows a price value in various fiat currencies to be reported to Exchange Coin.
  
• The Exchange Coin factors in the reported prices and runs a statistics algorithm that finds the current mean values.
  
• HST and bot trades could be filtered out statistically.

This would be organic and require trust of not a central reporting authority, but of all bitcoin users to volunteer honest reporting.

[Sultan]

I am guessing the decentralised nature of the exchange is with regards to the bids and asks alone, but as for the actual trade, that would probably be out of scope and dependant on the trust between the individuals on a trade.

I haven't installed or read the software, but it's how it works as far as I can tell.

Nevertheless, a decentralised database for, essentially storing the price of Bitcoin is crucial.

I didn't thought about this in this way. Thank you.
This sentence has really helped me.

I wrote this in another thread earlier today.

It's not like we need a full exchange that reports the volume of transactions. All we need is a general idea of the price bitcoin is being traded at.

• An Exchange Coin is created. The coins generated from it include pricing of all included crypto-currencies. More can be added if everyone agrees.
  
• Each coin client (including bitcoin) allows a price value in various fiat currencies to be reported to Exchange Coin.
  
• The Exchange Coin factors in the reported prices and runs a statistics algorithm that finds the current mean values.
  
• HST and bot trades could be filtered out statistically.

This would be organic and require trust of not a central reporting authority, but of all bitcoin users to volunteer honest reporting.

By the looks of it, Dark Exchange could be used to keep tabs on recent trading activity including volume, market depth etc, as all that would only require processing on the raw data. As long as the data is shared and verified by all members in the nodes, then the clients themselves can use this for whatever purpose.

[EhVedadoOAnonimato]

To decentralize dollars, you could make a coin backed by USDs:

What an interesting idea!

It's true that you would still have some almost-central "point of failures", which are the issuers, that could be shut down. But, if at least a few countries in the world have governments which are not that authoritarian, these issuers could establish in those places. Sort of what Pecunix tries to do in order to avoid the same fate of e-gold.

[cbeast]

To decentralize dollars, you could make a coin backed by USDs:

What an interesting idea!

It's true that you would still have some almost-central "point of failures", which are the issuers, that could be shut down. But, if at least a few countries in the world have governments which are not that authoritarian, these issuers could establish in those places. Sort of what Pecunix tries to do in order to avoid the same fate of e-gold.

Absolute genius! It's so simple why didn't anyone think of it. It would be as simple as writing a bill and have Congress vote on pinning a crypto-currency to USD. If you consider the rate of 21 Mil BTC created and if the US Treasury allowed the equivalent of $1 trillion to be created per year, then each bitcoin today would be worth over $400 each. It would be like a $2000 lottery every 10 minutes.

[jtimon]

To decentralize dollars, you could make a coin backed by USDs:

What an interesting idea!

It's true that you would still have some almost-central "point of failures", which are the issuers, that could be shut down. But, if at least a few countries in the world have governments which are not that authoritarian, these issuers could establish in those places. Sort of what Pecunix tries to do in order to avoid the same fate of e-gold.

Absolute genius! It's so simple why didn't anyone think of it. It would be as simple as writing a bill and have Congress vote on pinning a crypto-currency to USD. If you consider the rate of 21 Mil BTC created and if the US Treasury allowed the equivalent of $1 trillion to be created per year, then each bitcoin today would be worth over $400 each. It would be like a $2000 lottery every 10 minutes.

I think you have misunderstood the idea. 50 usdCoins won't be created each block. Only the issuing trust can create them. usdCoin miners would live on tx fees, but they could be merge mining.
You should read the beertoken proposal.

[EhVedadoOAnonimato]

usdCoin miners? A different chain? Why? Just "taint" satoshis, no need to create another chain, that's much more complicated.

EDIT: For those who don't know, one satoshi = 0,01µBTC, the minimal bitcoin unit. And "tainting" a coin is just marking it and following where it goes. The issuer could credit a bunch of addresses with 0,01µBTC each, and publicly state that each of those satoshis could be redeemable by, say, $1. They could be transferred and all just like any other bitcoin, the issuer himself cannot control where they go once they transfer it to someone else.

[cbeast]

usdCoin miners? A different chain? Why? Just "taint" satoshis, no need to create another chain, that's much more complicated.

EDIT: For those who don't know, one satoshi = 0,01µBTC, the minimal bitcoin unit. And "tainting" a coin is just marking it and following where it goes. The issuer could credit a bunch of addresses with 0,01µBTC each, and publicly state that each of those satoshis could be redeemable by, say, $1. They could be transferred and all just like any other bitcoin, the issuer himself cannot control where they go once they transfer it to someone else.

As long as they offer me a good fair price for my satoshis, I would turn them in for "tainting." Otherwise I'll stop mining and so will most others.

[Sultan]

Rather than create a new coin, I think payments should either be direct or in the architecture of Ripple. Certain nodes can act as bridges regarding cross-currency transactions.

[becoin]

I think payments should either be direct or in the architecture of Ripple.

This is the only path to follow if bitcoin exchanges continue to get their bank accounts canceled one after another.

[jtimon]

usdCoin miners? A different chain? Why? Just "taint" satoshis, no need to create another chain, that's much more complicated.

What's the problem with a new chain?
Another currency, another chain.
I don't think it's more complicated, the issuers can just use the beertoken code and change the name and backing of the currency.

It would be great to be able to trade ripple credits for bitcoins. But to make the trades atomic and decentralized you need:

1) Decentralized ripple (in progress).
2) commit scripts.

[EhVedadoOAnonimato]

What's the problem with a new chain?

There's a lot of infrastructure in bitcoin already, why reinvent the wheel? By tainting coins you can profit from everything that already exists, including miners. A different chain, even if this merged mining works, would have to offer incentives to miners... what does the issuer gain with that?
Using bitcoin only, you don't even need scripts or different rule sets concerning transactions, you can use simple, standard bitcoin transactions to make an atomic trade.

Well, I find it better to use what's already there, we have nothing to loose, only to gain. But it's true that both implementations could eventually work, so it's up to the issuers to choose their methods anyway.

[jtimon]

What's the problem with a new chain?

There's a lot of infrastructure in bitcoin already, why reinvent the wheel?

I really think an alternative chain is the simpler solution.
Maybe I don't understand your proposal.
What in the block chain says that the tainted satoshis are USD?
Do you need a centralized server from the issuers to tell you what satoshis are "tainted dollars"?

By tainting coins you can profit from everything that already exists, including miners. A different chain, even if this merged mining works, would have to offer incentives to miners...

Yes, fees. The same incentive that miners will ask for moving tainted coins. Merged mining already exists, is working for namecoin.

what does the issuer gain with that?

He doesn't need to take care of transactions and moves his currency within a system that can be trusted. He no longer needs to know the identity of his customers. The same that would be achieved with your proposal.

Using bitcoin only, you don't even need scripts or different rule sets concerning transactions, you can use simple, standard bitcoin transactions to make an atomic trade.

Well, I find it better to use what's already there, we have nothing to loose, only to gain. But it's true that both implementations could eventually work, so it's up to the issuers to choose their methods anyway.

Yes, is up to the issuers. But I think they will chose whatever is more convenient for the decentralized exchange, the bitcoin network and their currency itself. Your solution would not need replacements so it could be implemented earlier.
I hope someone implements one of the two.

[EhVedadoOAnonimato]

Maybe I don't understand your proposal.
What in the block chain says that the tainted satoshis are USD?
Do you need a centralized server from the issuers to tell you what satoshis are "tainted dollars"?

Yes, in a certain moment the server would announce new backed coins. After that, anyone can keep track of their movements. A common protocol could be defined so that different issuers can use the same software, just by adding different announcing servers.
Of course such announcement could be done in a P2P network, but there's no point in it, since the issuer is already a known central entity in which you have to trust anyway.

By tainting coins you can profit from everything that already exists, including miners. A different chain, even if this merged mining works, would have to offer incentives to miners...

Yes, fees. The same incentive that miners will ask for moving tainted coins. Merged mining already exists, is working for namecoin.

Do you know why people chose to create a new, independent chain for namecoin, with coins of its own?
Why not a chain dependent on bitcoin's chain, using bitcoin themselves for the transactions?

Back to the topic, the fee structure is already there for bitcoin transactions. By creating another chain, you'd need to pay fees for both wouldn't you?